CN-121030596-B - Power environment monitoring data visualization platform and abnormality diagnosis method

CN121030596BCN 121030596 BCN121030596 BCN 121030596BCN-121030596-B

Abstract

The application relates to the technical field of power environment data monitoring, and discloses a power environment monitoring data visualization platform and an abnormality diagnosis method, wherein the power environment monitoring data visualization platform comprises the steps of identifying abnormal data in power environment data; the method comprises the steps of constructing a causal heterogram of power environment data, exploring fault paths for each piece of abnormal data based on the causal heterogram, calculating fault confidence coefficient of each node in each fault path, screening and trimming each fault path based on the fault confidence coefficient of the node, acquiring weight values of the nodes based on the causal heterogram, calculating comprehensive causal strength of each fault path based on the fault confidence coefficient and the weight values of the nodes, and determining an actual fault path and diagnosing a fault source based on the comprehensive causal strength. The application improves the accuracy of the abnormal diagnosis result of the power environment monitoring system and the reliability of fault source positioning.

Inventors

ZHANG LIJUN
ZHANG XINGHUI
QU HAIYANG
WANG LEYAO

Assignees

江苏奋为信息技术有限公司

Dates

Publication Date: 20260512
Application Date: 20250805

Claims (9)

1. A power environment monitoring data abnormality diagnosis method is characterized by comprising the following steps: The abnormal data in the dynamic environment data is identified, and a causal heterogram of the dynamic environment data is constructed; exploring a fault path for each item of abnormal data based on the causal disparity map; calculating the fault confidence coefficient of each node in each fault path, and carrying out linkage consistency verification and compensation correction on the fault confidence coefficient of the node; screening and trimming each fault path based on the fault confidence of the node; Acquiring a weight value of a node based on the causal heterogram, and calculating the comprehensive causal strength of each fault path based on the fault confidence and the weight value of the node; Determining an actual fault path and diagnosing a fault source based on the integrated causal strength; the identifying the abnormal data in the power environment data specifically comprises the following steps: respectively establishing a time sequence of each item of power environment data, extracting the latest m monitoring values in the time sequence for any item of power environment data, and calculating the change rate and fluctuation degree of the power environment data based on the latest m monitoring values, wherein m is a positive integer; setting a reference value interval, a change rate threshold value and a fluctuation degree threshold value for each dynamic environment data respectively; if the power environment data meet at least one of the change rate greater than the change rate threshold, the fluctuation degree greater than the fluctuation degree threshold and at least n monitoring values out of the m monitoring values exceeding a reference value interval, the power environment data are abnormal data, wherein n is a positive integer less than or equal to m; the method for calculating the fault confidence coefficient of each node in each fault path is to assign an initial value to the fault confidence coefficient based on the time sequence of the power environment data corresponding to the node, and concretely comprises the steps that if the power environment data corresponding to the node is not abnormal data, the fault confidence coefficient is 0, otherwise: The method comprises the steps of calculating the average value of the latest m monitoring values in a time sequence to be used as an observation value of a node, acquiring the change rate and the fluctuation degree calculated based on the latest m monitoring values, and giving an initial value to fault confidence based on any one of the absolute value of the difference between the observation value and the midpoint of a reference value interval, the difference between the change rate and a change rate threshold value and the difference between the fluctuation degree and a fluctuation degree threshold value; And if any non-target node fails the abnormal direction verification or the abnormal hysteresis verification, the confidence coefficient of the corresponding non-target node is set to be 0.
2. The method for diagnosing the abnormality of the power environment monitoring data according to claim 1, wherein the power environment data comprises power data and environment data, wherein the power data comprises operation parameters of power supply equipment in a power system; Partitioning the power supply equipment and the working environment thereof according to the functional area or the physical position, and independently recording each power environment data of each partition as one power environment data.
3. The method for diagnosing an abnormality of power environment monitoring data according to claim 2, wherein the causal graph comprises nodes and directed edges, wherein any node corresponds to one piece of power environment data, and wherein any directed edge is used for connecting two nodes and is directed from an upstream node to a downstream node; if any two nodes are connected through a directed edge, the corresponding two nodes have abnormal linkage relation, and the method specifically comprises the steps that when power environment data corresponding to an upstream node in the corresponding two nodes are abnormal, probability is given to the power environment data corresponding to a downstream node, and the probability is equal to the edge weight of the corresponding directed edge; And collecting historical data, counting the probability of any power environment data abnormality except for the probability of any power environment data abnormality when any power environment data abnormality occurs, and assigning a value to a weight value of a corresponding directed edge in the causal heterogram.
4. The method for diagnosing the abnormality of the power environment monitoring data according to claim 3, wherein the fault path comprises nodes and directed edges in a causal heterogram, the nodes corresponding to the corresponding abnormal data are marked as target nodes when the fault path is explored for any abnormal data, any fault path is loop-free and branch-free, and the target nodes have no downstream nodes; The fault path exploration method includes the steps of exploring nodes in the direction of the opposite directional edges in a causal heterogram by taking a target node as a starting point, updating the current nodes and the opposite directional edges of the fault path when one node is explored, calculating the cumulative products of the weight values of all the current directed edges of the fault path, and stopping exploration to obtain a fault path if the cumulative products of the weight values are smaller than a preset minimum weight threshold value.
5. The method for diagnosing the power environment monitoring data abnormality according to claim 4, wherein the verification of the abnormality direction comprises the steps of respectively extracting the difference between the observed value of any non-target node and the adjacent downstream node in the fault path and the midpoint of the reference value interval, and identifying the abnormality direction of the non-target node and the adjacent downstream node; The abnormal linkage direction comprises positive linkage and negative linkage, wherein the positive linkage indicates that the abnormal direction of the non-target node is the same as that of the adjacent downstream node, and the negative linkage indicates that the abnormal direction is opposite; if the abnormal direction of the non-target node and the adjacent downstream node does not meet the abnormal linkage direction, the non-target node fails to pass the abnormal direction verification.
6. The method for diagnosing the abnormality of the power environment monitoring data according to claim 5, wherein the verification of the abnormality hysteresis specifically comprises intercepting a time sequence corresponding to each node in a fault chain through a sliding window with a length of m monitoring values, judging whether the power environment data is abnormal data after each sliding interception; the method comprises the steps of calculating the time difference between abnormal starting points of any non-target node and adjacent downstream nodes in a fault path to serve as abnormal lag time of the corresponding non-target node, setting a lag time interval for each directed edge in a causal difference graph, and if the abnormal lag time of the non-target node is not located in the lag time interval of the directed edge between the non-target node and the adjacent downstream nodes, failing to pass the abnormal lag verification of the non-target node.
7. The method for diagnosing the abnormality of the power environment monitoring data according to claim 6, wherein the screening and trimming of each fault path comprises the steps of carrying out primary screening and primary trimming on the fault path after giving an initial value to the fault confidence coefficient of each node in each fault path; the method for carrying out primary screening or secondary screening on the fault path comprises the following steps: If the fault confidence coefficient of all non-target nodes except the target node in any fault path is 0, deleting the corresponding fault path; If any fault path comprises at least one non-target node with the fault confidence coefficient not being 0, and a non-target node with the confidence coefficient being 0 exists between at least one of the non-target nodes with the fault confidence coefficient not being 0 in the fault path and the target node, deleting the corresponding fault path; the method for performing primary trimming or secondary trimming on the fault path is as follows: If any fault path comprises at least one non-target node with the fault confidence coefficient not being 0, and no non-target node with the confidence coefficient being 0 exists between any non-target node with the fault confidence coefficient not being 0 and the target node in the fault path, trimming the corresponding fault path, and specifically comprises deleting all non-target nodes with the fault confidence coefficient being 0 in the fault path.
8. The method for diagnosing the abnormality of the power environment monitoring data according to claim 7, wherein the weight value of any node is the weight value of a directed edge between the corresponding node and the adjacent downstream node in the corresponding fault path, the calculating of the comprehensive causal strength of any fault path is based on the fault confidence coefficient and the weight value of the node, and the method specifically comprises the steps of multiplying the fault confidence coefficient of each node by the corresponding weight value to obtain the weighted confidence coefficient of the node; the method for determining the actual fault path comprises the steps of sequencing the fault paths according to the magnitude of the comprehensive causal strength, and selecting the fault path with the maximum comprehensive causal strength as the actual fault path; The method for diagnosing the fault source comprises the following steps of marking a first node in the actual fault path as the fault source, wherein the first node is a node without an upstream node in the actual fault path.
9. The power environment monitoring data visualization platform is used for realizing the power environment monitoring data abnormality diagnosis method according to any one of claims 1-8, and is characterized by comprising an abnormality identification module, a causal modeling module, a causal reasoning module, a path evaluation module and a visualization module, wherein: the abnormality identification module is used for identifying abnormal data in the power environment data; the causal modeling module is used for constructing a causal heterogram of dynamic environment data; The causal reasoning module explores fault paths for each piece of abnormal data based on the causal difference map, and screens and prunes each fault path; The path evaluation module is used for calculating the comprehensive causal strength of each fault path, determining the actual fault path based on the comprehensive causal strength and diagnosing the fault source; The visualization module is used for performing visual display on power environment data, a causal heterogram, an actual fault path and a fault source.

Description

Power environment monitoring data visualization platform and abnormality diagnosis method Technical Field The application relates to the technical field of power environment data monitoring, in particular to a power environment monitoring data visualization platform and an abnormality diagnosis method. Background In order to realize real-time state monitoring and risk early warning of a power environment system, the power environment monitoring system is widely adopted in the industry, various power supply parameters and environment indexes are continuously monitored by utilizing a sensor, acquisition equipment and a data analysis platform, the state change of the power supply equipment and the running environment thereof can be reflected to a certain extent, and the potential risk can be found by assisting operation and maintenance personnel in time. However, most existing power environment monitoring systems stay at a level based on threshold judgment or single point index anomaly detection, and have some disadvantages. Existing monitoring systems lack in-depth modeling of complex linkage relationships between devices and environments. The running state of the power equipment and environmental factors have a high coupling relation, the prior art often carries out isolated monitoring on various parameters, lacks systematic causal logic reasoning, cannot accurately judge the relevance and linkage between different parameter changes, and causes the root cause of abnormal events to be difficult to effectively identify. The traditional fault diagnosis method is mostly based on a single reasoning path or fixed diagnosis rules, and lacks multi-path and multi-result parallel reasoning capability. In an actual system, the same abnormal phenomenon can be caused by multiple reasons, a complex causal transmission chain exists between different fault sources, a single reasoning conclusion easily causes misjudgment, the real fault state of the system cannot be comprehensively and accurately reflected, and the accuracy and the comprehensiveness of fault source positioning are limited. The Chinese patent with the authority bulletin number of CN110046074B discloses a data center power environment monitoring system and a data center power environment monitoring method, wherein the power environment monitoring operation is executed through a set acquisition module and a monitoring module, the acquisition module is used for acquiring data of a data center power system by using a sensor deployed on the acquisition module and sending the acquired data to the monitoring module, the monitoring module is used for receiving the acquired data sent by the acquisition module, establishing a data analysis model for the acquired data, analyzing the acquired data by using the data analysis model, identifying whether the data center power environment system is abnormal or not, and simultaneously carrying out safety precaution on the data center power environment system, thereby achieving the purpose of monitoring the equipment operation condition and the environment state of the data center power system in real time, improving the safety of data storage, and meanwhile, the monitoring system can be checked through APP, and can also be used for calling information expected to be checked at any time and any place, thereby improving the convenience of the monitoring system. However, the technical scheme still has the problem of lacking multi-path and multi-result parallel reasoning capability in the background technology of the application. The information disclosed in this background section is only for enhancement of understanding of the general background of the application and should not be taken as an acknowledgement or any form of suggestion that this information forms the prior art already known to a person of ordinary skill in the art. Disclosure of Invention The technical problem to be solved by the application is to overcome the defects of the prior art, provide a power environment monitoring data visualization platform and an abnormality diagnosis method, and improve the accuracy of an abnormality diagnosis result of a power environment monitoring system and the reliability of fault source positioning. In order to solve the technical problems, the application provides the following technical scheme: in one aspect, the application provides a method for diagnosing abnormality of power environment monitoring data, comprising the following steps: The abnormal data in the dynamic environment data is identified, and a causal heterogram of the dynamic environment data is constructed; exploring a fault path for each item of abnormal data based on the causal disparity map; calculating the fault confidence coefficient of each node in each fault path, and carrying out linkage consistency verification and compensation correction on the fault confidence coefficient of the node; screening and trimming each fault path based on the fault confiden